Abstract
The chemokine receptor CXCR4 is a G protein-coupled receptor that plays an important role in several biological processes, such as trafficking and homeostasis of immune cells (like T lymphocytes), alteration of cell skeleton rearrangement and cell migration. To investigate whether the CXCR4 protein impacts on lung cancer prognosis, a meta-analysis was performed. Our meta-analysis study involved 2,037 lung cancer patients from 24 studies by a comprehensive search from PubMed, Embase and CNKI databases up to September 2014. Odds ratio (OR) or hazard ratio (HR) with 95% confidence interval (CI) were used to evaluate the relationship. We found that the CXCR4 expression was significantly associated with lymph node metastasis (OR = 3.79, 95% CI: 2.15-6.68), distant metastasis (OR = 3.67, 95% CI: 1.84-7.32), tumor stage (OR = 2.78, 95% CI: 1.77-4.39) and overall survival (HR = 1.63, 95% CI: 1.16-2.30). In conclusion, CXCR4 might be a new prognostic biomarker, and it might become a new diagnosis and therapeutic target in lung cancer.
Keywords: CXCR4, lung cancer, prognosis, meta-analysis
Introduction
Lung cancer is a high malignant carcinoma and it has been reported to be the first leading cause of cancer death in the United States [1]. Despite the advanced diagnostic techniques for early detection of lung cancer, the prognosis of lung cancer patients is still unsatisfactory. Even in early stage lung cancer, many patients developed recurrent disease and died of metastasis [2]. Because of the limited knowledge of lung cancer and technology for treatment, lung cancer can hardly be cured as our expectation. So it is necessary to explore prognostic factors to predict the outcomes of lung cancer patients, which can guide doctors to making effective strategies and increasing survival time for patients.
Chemokines are a small molecules family that adjusts immune responses. They are divided into two types, namely CXC and CC, by the position of the first two cysteines in their sequence [3]. The chemokine receptor CXCR4 is a G protein-coupled receptor (GPCR) that binds its ligand stromal cell-derived factor 1 (SDF-1, also known as CXCL12). CXCR4 has been identified to play an important role in several biological processes, such as trafficking and homeostasis of immune cells (like T lymphocytes) [4], leading to alteration of cell skeleton rearrangement and cell migration [5]. In several types of cancer, CXCR4 also contributes to neoplasia and the development of cancer [6]. Recent studies has reported that CXCR4 was related to cell survival, differentiation, proliferation and metastasis in breast cancer [7], colorectal cancer [8], gastric cancer [9], prostate cancer [10], etc. These studies derived that CXCR4 is merging as attractive targets for developing novel prognostic approaches for cancers.
Since Spano et al. first identified the relationship between CXCR4 expression and lung cancer patient’s prognosis [11]; several studies have been published to describe this association [12-34]. But each of the studies has failed to provide conclusive results. Our present meta-analysis study was conducted to quantitatively and precisely estimate the potential effect of CXCR4 and lung cancer prognosis.
Materials and methods
Publication search
We searched published studies in the PubMed, Embase and CNKI databases updated to September 2014. The search was limited by using the following search terms: (CXCR4 OR chemokine receptor 4) AND (lung OR pulmonary) AND (cancer OR neoplasms OR carcinoma OR tumor OR adenocarcinoma) AND prognosis. Furthermore, reference lists of main reports and review articles were also reviewed to identify additional relevant publications.
Selection criteria
Two authors reviewed the retrieved titles and abstracts to determine the eligibility of the studies for inclusion in our meta-analysis independently. Published studies were included based on the following criteria: (1) patients with distinctive lung cancer diagnosis by pathology; (2) CXCR4 expression was detected by immunohistochemistry or RT-PCR; (3) CXCR4 expression on human lung cancer tissue; (4) the main outcome of interest focus on prognostic factors and clinicopathological features; (5) full length paper with sufficient data to calculate the odds ratio (OR) or hazard ratio (HR) estimates and their 95% confidence intervals (95% CI). We excluded studies with the following criteria: (1) articles about cell lines or animals; (2) CXCR4 expression on peripheral blood; (3) studies without sufficient data on prognostic factors or clinicopathological features; (4) reviews without original data and studies with duplicated data.
Data extraction
Two investigators performed the data evaluation independently. The following data were extracted from each study: first author, year, country, patients, method, antibody, subcellular localization, No. of patients (CXCR4 high/low), duration of follow-up, prognostic factors (age, gender, tumor size, differentiation, smoking, T stage, N stage, M stage, TNM stage) and survival (overall survival (OS) and disease-free survival (DFS)).
Data synthesis and statistical analysis
All of prognostic factors were analyzed as dichotomous variables; these data were analyzed by random-effect method, and were measured in OR with 95% CI. In survival analysis, the data were measured in HR with 95% CI. If the HR or standard errors (SEs) were not reported in included studies, we calculate or estimate the HR from available data or Kaplan-Meier curves using the methods reported by Tierney et al. [35]. Statistical heterogeneity was estimated by means of Cochran’s Q test and-squared test. Their-squared test represents the percentage of variation to heterogeneity, which is categorized as low (0-40%), moderate (40-60%), high (60-90%), very high (> 90%). Subgroup analyses were carried out based on geographic location, types of cancer or staining location of included studies if a significant heterogeneity was found in overall meta-analysis. Sensitivity analyses were performed by omitting one study at a time to check if the inclusion criteria affected the final results. To identify any potential publication bias, we used Begg’s test and Egger’s test, and only showed Begg’s test in Figures. All statistical analyses were performed with Review Manager 5.2 and STATA 12.0.
Results
Systematic review
We identified 251 studies that fit our search strategy, but only 38 studies matched with inclusion criteria and content (Figure 1). After reviewing full text, 2 study was review articles, 5 study was excluded because of ineligible study object, 4 studies were excluded because they were lack of sufficient information to calculate effect estimates, 3 studies were excluded because duplicate report on the same population. Finally, we identified 24 studies to analysis [11-34].
Figure 1.
The flow diagram of included/excluded studies.
Detailed characteristics of these studies were provided in Table 1. The included studies were published between 2004 and 2014, and included 2037 lung cancer patients. 17 studies were performed in China, 3 studies in Japan, 2 studies in United States, each 1 study in France and Canada. In prognostic factors, 11 studies were identified the relationship between age and lung cancer prognosis, 17 studies about gender, 10 studies about differentiation, 6 studies about tumor size, 4 studies about smoking, 5 studies about T stage, 16 studies about N stage, 9 studies about M stage, 15 studies about TNM stage. In survival analysis, 10 studies were demonstrated the association between OS and lung cancer prognosis and 4 studies about DFS.
Table 1.
Characteristics of studies included in the meta-analysis
| First author [reference] | Year | Country | Patients | Method | Antibody | Subcellular localization | No. of patients (CXCR4 high/low) | Duration of follow-up (months) | Prognostic factors | Survival |
|---|---|---|---|---|---|---|---|---|---|---|
| Spano [11] | 2004 | France | Stage I NSCLS | IHC | Abcam | Cytoplasm, nucleus | 61 (17/44) | 144 | A G Sm T M | NA |
| Zhang [12] | 2006 | China | NSCLC | IHC | Santa Cruz | Cytoplasm, nucleus | 72 (46/26) | 60 | G T N M St | NA |
| Cai [13] | 2006 | China | NSCLC | IHC | Wuhan Boster | Cytoplasm, membrane | 40 (18/22) | NA | A G D N | NA |
| Suzuki [14] | 2008 | Japan | NSCLC | IHC | Santa Cruz | NA | 90 (22/68) | 120 | NA | OS |
| Wagner [15] | 2009 | United States | NSCLC | IHC | R&D | Cytoplasm, nucleus | 154 (62/92) | 180 | G T N M St | DFS |
| Iwakiri [16] | 2009 | Japan | NSCLC | RT-PCR | NA | NA | 79 (40/39) | 60 | NA | DFS OS |
| Reckamp [17] | 2009 | United States | NSCLC | IHC | R&D | NA | 16 (5/11) | 34 | G Sm St | OS |
| Xiao [18] | 2009 | China | Lung cancer | IHC | Wuhan Boster | Cytoplasm, membrane | 82 (42/40) | NA | A G Si D N St | NA |
| Minamiya [19] | 2010 | Japan | AD of the lung | IHC | Leinco | Cytoplasm, nucleus | 79 (37/42) | 60 | G D Si N St | DFS OS |
| Yao [20] | 2010 | China | NSCLC | IHC | Abcam | NA | 52 (33/19) | NA | A G D N St | NA |
| Chen [21] | 2011 | China | NSCLC | IHC | Abcam | Cytoplasm, membrane | 64 (51/13) | NA | M | NA |
| Otsuka [22] | 2011 | Canada | Stage IV NSCLC | IHC | UMB2 | Cytoplasm | 170 (29/141) | 50 | G Sm M | OS |
| Wang [23] | 2011 | China | NSCLC | IHC | R&D | Cytoplasm | 208 (117/91) | 70 | A G Si D Sm T N St | OS |
| Xi [24] | 2011 | China | NSCLC | IHC | Wuhan Boster | NA | 62 (19/43) | NA | Si D N St | NA |
| Li [25] | 2012 | China | SCLC | IHC | Abcam | Cytoplasm | 65 (31/34) | 87 | N M St | DFS |
| Zhou [26] | 2012 | China | Stage III NSCLC | IHC | Boao Sen | NA | 105 (72/33) | NA | M | NA |
| Geng [27] | 2012 | China | NSCLC | IHC | Wuhan Boster | NA | 95 (40/55) | NA | A G Si D N St | NA |
| Hu [28] | 2012 | China | NSCLC | IHC | ZGBBT | NA | 75 (60/15) | 60 | D T N | NA |
| Wang [29] | 2012 | China | Lung cancer | IHC | NA | NA | 72 (24/48) | NA | A G N St | NA |
| Zobair [30] | 2013 | China | NSCLC | IHC | Abcam | Cytoplasm, nucleus | 125 (62/63) | 45 | G M St | OS |
| Wang [31] | 2013 | China | NSCLC | IHC | Abcam | Cytoplasm | 86 (53/33) | NA | A G D N St | NA |
| Li [32] | 2014 | China | SCLC | IHC | R&D | Cytoplasm, membrane | 50 (35/15) | 60 | A G Si N St | OS |
| Liu [33] | 2014 | China | NSCLC | IHC | Abcam | Membrane, nucleus | 75 (45/30) | 40 | A G N M | NA |
| Ji [34] | 2014 | China | NSCLC | IHC | Wuhan Boster | Cytoplasm | 60 (47/13) | NA | A G D N St | NA |
A, age; G, gender; Sm, smoke; Si, tumor size; D, differentiation; T, T stage; N, N stage; M, M stage; St, tumor stage; OS, overall survival; DFS, disease-free survival; NA, not available; AD, adenocarcinoma.
Association of CXCR4 expression with prognosis factors
CXCR4 expression was not significant related to prognosis factors, such as age (old patients vs. young patients) (pooled OR = 0.97, 95% CI: 0.71-1.33, I2 = 16.1%), gender (male vs. female) (pooled OR = 1.05, 95% CI: 0.84-1.32, I2 = 0.0%), tumor size (large vs. small) (pooled OR = 0.90, 95% CI: 0.54-1.48, I2 = 40%), differentiation (poor vs. moderate and high) (pooled OR = 1.31, 95% CI: 0.75-2.30, I2 = 62.8%), smoking (long time vs. short time or never) (pooled OR = 1.22, 95% CI: 0.55-2.69, I2 = 29.7%), T stage (T3, 4 vs. T1, 2) (pooled OR = 1.88, 95% CI: 0.76-4.64, I2 = 71.1%) (Table 2).
Table 2.
Meta-analysis with a random-effect model for the association of CXCR4 expression and prognosis factors
| Categories | No. of studies | Comparison | Pooled OR (95% CIs) | I-squared value | ph * |
|---|---|---|---|---|---|
| Age | 11 | Old vs. young | 0.97 (0.71, 1.33) | 16.1% | 0.291 |
| Gender | 17 | Male vs. female | 1.05 (0.84, 1.32) | 0.0% | 0.891 |
| Tumor size | 6 | Large vs. small | 0.90 (0.54, 1.48) | 40.0% | 0.139 |
| Differentiation | 10 | Poor vs. moderate and high | 1.31 (0.75, 2.30) | 62.8% | 0.004 |
| Smoking | 4 | Long time vs. short time or never | 1.22 (0.55, 2.69) | 29.7% | 0.234 |
| T stage | 5 | T3, 4 vs. T1, 2 | 1.88 (0.76, 4.64) | 71.1% | 0.008 |
| N stage | 16 | N1, 2 vs. N0 | 3.79 (2.15, 6.68) | 76.6% | 0.000 |
| M stage | 9 | M1 vs. M0 | 3.67 (1.84, 7.32) | 68.9% | 0.001 |
| Tumor stage | 15 | III, IV vs. I, II | 2.78 (1.77, 4.39) | 62.0% | 0.001 |
P value for heterogeneity of each meta-analysis.
However, CXCR4 expression correlated to some prognosis factors, such as N stage (N1, 2 vs. N0) (pooled OR = 3.79, 95% CI: 2.15-6.68, I2 = 76.6%), M stage (M1 vs. M0) (pooled OR = 3.67, 95% CI: 1.84-7.32, I2 = 68.9%), tumor stage (III, IV vs. I, II) (pooled OR = 2.78, 95% CI: 1.77-4.39, I2 = 62.0%) (Table 2).
CXCR4 expression on lung cancer survival
Eight studies investigating OS and 4 studies identifying DFS were pooled into the meta-analysis. CXCR4 positive expression significantly correlated with poor OS (pooled HR = 1.63, 95% CI: 1.16-2.30, I2 = 54.9%). However, CXCR4 expression was not related to DFS (pooled HR = 1.42, 95% CI: 0.65-3.09, I2 = 74.1%) (Figure 2).
Figure 2.
Meta-analysis with a random-effect model for the association of CXCR4 expression and survival factors such as overall survival (A) and disease-free survival (B).
Subgroup analyses
We take subgroup analyses in meta-analysis with relative high heterogeneity (I-square > 40%). In subgroup analyses, studies were stratified by geographic location (Asia, Europe and North America), types of cancer (NSCLC, SCLC and Lung cancer) or staining location (cytoplasm & nucleus, cytoplasm & membrane, nucleus & membrane and cytoplasm). In addition, heterogeneity was also showed in the studies which adjusted for these aforementioned risk factors (Table 3).
Table 3.
Subgroup analyses
| No. of studies | Pooled OR/HR (95% CIs) | I-squared value | ph * | |
|---|---|---|---|---|
| Differentiation | ||||
| Overall | 10 | 1.31 (0.75, 2.30) | 62.8% | 0.004 |
| Geographic location | ||||
| Asia | 10 | 1.31 (0.75, 2.30) | 62.8% | 0.004 |
| Types of cancer | ||||
| NSCLC | 8 | 1.18 (0.58, 2.39) | 69.4% | 0.002 |
| Lung cancer | 2 | 1.84 (0.91, 3.72) | 0.0% | 0.780 |
| Staining location | ||||
| Cytoplasm & nucleus | 2 | 0.22 (0.02, 1.99) | 70.9% | 0.064 |
| Cytoplasm & membrane | 1 | 1.71 (0.72, 4.06) | - | - |
| Cytoplasm | 2 | 1.39 (0.82, 2.38) | 0.0% | 0.774 |
| T stage | ||||
| Overall | 5 | 1.88 (0.76, 4.64) | 71.1% | 0.008 |
| Geographic location | ||||
| Asia | 3 | 3.01 (1.07, 8.45) | 55.5% | 0.106 |
| Europe | 1 | 1.83 (0.58, 5.83) | - | - |
| North America | 1 | 0.57 (0.23, 1.40) | - | - |
| Types of cancer | ||||
| NSCLC | 5 | 1.88 (0.76, 4.64) | 71.1% | 0.008 |
| Staining location | ||||
| Cytoplasm & nucleus | 3 | 0.99 (0.49, 1.97) | 23.9% | 0.269 |
| Cytoplasm | 1 | 3.92 (1.88, 8.19) | - | - |
| N stage | ||||
| Overall | 16 | 3.79 (2.15, 6.68) | 76.6% | 0.000 |
| Geographic location | ||||
| Asia | 15 | 4.26 (2.45, 7.41) | 71.5% | 0.000 |
| North America | 1 | 0.80 (0.39, 1.64) | - | - |
| Types of cancer | ||||
| NSCLC | 12 | 2.87 (1.66, 4.94) | 68.0% | 0.000 |
| SCLC | 2 | 2.37 (1.06, 5.30) | 0.0% | 0.922 |
| Lung cancer | 2 | 31.15 (12.75, 76.10) | 0.0% | 0.341 |
| Staining location | ||||
| Cytoplasm & nucleus | 3 | 1.21 (0.23, 6.21) | 88.3% | 0.000 |
| Cytoplasm & membrane | 3 | 7.40 (1.52, 35.96) | 77.5% | 0.012 |
| Nucleus & membrane | 1 | 6.83 (1.94, 24.09) | - | - |
| Cytoplasm | 4 | 2.64 (1.70, 4.11) | 0.0% | 0.635 |
| M stage | ||||
| Overall | 9 | 3.67 (1.84, 7.32) | 68.9% | 0.001 |
| Geographic location | ||||
| Asia | 6 | 5.67 (3.62, 8.86) | 0.0% | 0.710 |
| Europe | 1 | 0.59 (0.17, 2.14) | - | - |
| North America | 2 | 2.75 (0.20, 37.46) | 80.3% | 0.024 |
| Types of cancer | ||||
| NSCLC | 8 | 3.60 (1.64, 7.94) | 72.6% | 0.001 |
| SCLC | 1 | 4.39 (1.55, 12.43) | - | - |
| Staining location | ||||
| Cytoplasm & nucleus | 4 | 3.44 (0.98, 12.06) | 73.6% | 0.010 |
| Cytoplasm & membrane | 1 | 4.39 (1.08, 17.89) | - | - |
| Nucleus & membrane | 1 | 7.00 (2.49, 19.70) | - | - |
| Cytoplasm | 2 | 1.91 (0.40, 9.24) | 81.5% | 0.020 |
| Tumor stage | ||||
| Overall | 15 | 2.78 (1.77, 4.39) | 62.0% | 0.001 |
| Geographic location | ||||
| Asia | 13 | 3.19 (2.12, 4.80) | 47.5% | 0.029 |
| North America | 2 | 0.78 (0.39, 1.57) | 0.0% | 0.411 |
| Types of cancer | ||||
| NSCLC | 11 | 2.68 (1.44, 4.96) | 71.0% | 0.000 |
| SCLC | 2 | 2.35 (1.07, 5.20) | 0.0% | 0.545 |
| Lung cancer | 2 | 3.54 (1.73, 7.21) | 9.1% | 0.294 |
| Staining location | ||||
| Cytoplasm & nucleus | 4 | 1.37 (0.45, 4.19) | 82.3% | 0.001 |
| Cytoplasm & membrane | 4 | 2.87 (1.73, 4.76) | 0.0% | 0.717 |
| Cytoplasm | 3 | 6.36 (3.32, 12.19) | 1.7% | 0.362 |
| OS | ||||
| Overall | 8 | 1.63 (1.16, 2.30) | 54.9% | 0.030 |
| Geographic location | ||||
| Asia | 6 | 1.51 (0.97, 2.36) | 64.3% | 0.016 |
| North America | 2 | 1.99 (0.96, 4.11) | 33.8% | 0.219 |
| Types of cancer | ||||
| NSCLC | 8 | 1.63 (1.16, 2.30) | 54.9% | 0.030 |
| Staining location | ||||
| Cytoplasm & nucleus | 2 | 0.91 (0.16, 5.31) | 91.4% | 0.001 |
| Cytoplasm | 3 | 1.83 (1.40, 2.39) | 0.0% | 0.725 |
| DFS | ||||
| Overall | 4 | 1.42 (0.65, 3.09) | 74.1% | 0.009 |
| Geographic location | ||||
| Asia | 3 | 1.09 (0.40, 2.99) | 76.4% | 0.014 |
| North America | 1 | 2.80 (1.39, 5.65) | - | - |
| Types of cancer | ||||
| NSCLC | 3 | 1.22 (0.37, 4.08) | 81.9% | 0.004 |
| SCLC | 1 | 1.94 (1.08, 3.48) | - | - |
| Staining location | ||||
| Cytoplasm & nucleus | 2 | 1.04 (0.14, 7.72) | 90.9% | 0.001 |
| Cytoplasm | 1 | 1.94 (1.08, 3.48) | - | - |
P value for heterogeneity of each meta-analysis.
Sensitivity analyses
In sensitivity analyses, we sequentially removed one study at a time and re-analyzed the data to explore the origin of the heterogeneity. This showed that the study by Minamiya et al. [19] substantially impacted the pooled HR in OS and DFS meta-analysis (Figure 3). After omitting this study, heterogeneity was no longer observed in OS (pooled HR = 1.89, 95% CI: 1.52-2.34, I2 = 0.0%) and DFS (pooled HR = 2.14, 95% CI: 1.42-3.22, I2 = 0.0%). In other sensitivity analyses, we found that no single study altered the original results or heterogeneity significantly.
Figure 3.
Influence of individual studies on the pooled HR in meta-analysis of overall survival (A) and disease-free survival (B).
Publication bias
Begg’s and Egger’s were created for assessment of possible publication bias. Both of them suggested that the publication bias had little influence on this meta-analysis results (P > 0.05). We only showed Begg’s test in Figures 4 and 5.
Figure 4.
Begg’s test results of CXCR4 and prognosis factors such as age (A), gender (B), tumor size (C), differentiation (D), smoking (E), T stage (F), N stage (G), M stage (H), tumor stage (I).
Figure 5.
Begg’s test results of the overall survival (A) and disease-free survival (B).
Discussion
Taking statistics from the USA as an example of the evolution of lung cancer [1], it is clear that the prognostic factor for lung cancer which prevents and cures this type of cancer needs to be improved. Exploring new molecular biological prognostic and predictive markers is a hot topic in modern medicine. In recent studies, CXCR4 was considered to be a new prognostic marker in several types of cancer [36-38]. However, results of the relationship between CXCR4 expression and lung cancer prognosis are not conformable in several studies [11-34]. To our knowledge, this meta-analysis is the first study to systematically evaluate the relationship between CXCR4 expression and lung cancer prognosis.
In the present study, a combined analysis of 24 articles which showed the detection of the CXCR4 expression in tumor tissues with poor prognosis outcome in lung cancer patients who were detected with high level of CXCR4 expression. Our meta-analysis results indicated that CXCR4 expression was significantly correlated to lymph node metastasis (N stage), distant metastasis (M stage), tumor stage and overall survival. CXCR4 overexpression also shorten disease-free survival but not notable. On the other hand, high level of CXCR4 expression was also found in patients like elderly, male, smoking, and patients with low differentiate, small size tumor or deep invasion, but none of these results showed any significance.
What makes CXCR4 relate to the poor prognosis among lung cancer patients? As far as we know, different types of cancers express different chemokines and their receptors, but CXCR4 is the only one that expresses in the majority of cancer types [39,40]. After Kijima et al. first reported the CXCR4 expression in lung cancer [41]; many studies were conducted to investigate the relationship between CXCR4 expression and lung cancer prognosis. Interestingly, Minamiya et al. found that CXCR4 represented low level or negative in normal lung cells while high expression was observed in lung cancer cells [19]. CXCR4 and its chemokine ligand 12 (CXCL12) recruit the endothelial progenitor cells into tumors indirectly, and result in neoangiogenesis [42]. Meantime, reports showed that the EGF receptor HER2 increases CXCR4 expression as well as the invasion and metastasis of HER2-positive breast cancer cells [43]. Lung cancer expressed a substantial percentage of EGFRs on cell surface, and the association between EGFR and CXCR4 may also exist in lung cancer as described in Zobair’s study [30]. It is well known that hypoxia is a common phenomenon occurring in the majority of human tumors and has been proved to play an important role in tumor progression. In Liu’s study, hypoxia can regulate the CXCR4 mediated metastasis and HIF-dependent invasion, migration and adhesion [44]. From our results, we also found that CXCR4 expression was correlated to lymph node metastasis and distant metastasis, which inferred that CXCR4 was associated with tumor microenvironment and enhanced the cancer cell survival. Bertolini et al. Found that CXCR4 was related to lung cancer progenitor cells, and this was the first in vivo evidence for a tumorigenic and metastatic subpopulation in lung cancer which characterized by CD133+ expression together with CXCR4+ [45]. CXCR4 was also reported to be associated with cancer stem cells, and in vivo study showed its encouraging effect on the chemo resistance [45]. Furthermore, CXCR4 activation augmented the signaling pathways related to cell survival and growth, such as MAPK [46] and PI3K signaling pathways in lung cancer cells [41]. Therefore, the molecular biological mechanisms of how CXCR4 overexpression affects the lung cancer prognosis are complicated and still needs more exploration. For the first time, our meta-analysis study revealed that CXCR4 could be a potential biomarker for poor prognosis of lung cancer.
This meta-analysis had several strengths. It included total 2037 lung cancer patients which should provided sufficient statistical power to detect the association between CXCR4 expression and lung cancer prognosis. A further strength was that we carried out subgroup and sensitivity analyses to explore the potential sources of heterogeneity.
Some limitations of this meta-analysis should be pointed out. First, all published studies and papers written in English or Chinese, Some related published or unpublished studies that meet the inclusion criteria were missed. Most of the studies reported positive results, and studies of negative results were all rejected. Second, the two different types of lung cancer (NSCLC and SCLC) might have different biological behaviors. In this meta-analysis, only two studies demonstrated to SCLC, and only one study could successfully evaluated estimate HR or OR in each prognosis factors except N stage and tumor stage. Third, most of the included studies had data of the CXCR4 expression which were detected by IHC methods. It might have some bias because of different antibodies and different standards of positive/negative CXCR4 expression. However, it was not available for us to take subgroup analysis to analyze the underlying bias of IHC on the pooled ORs or HRs. Forth, the data of overall survival and disease-free survival was not performed by multivariate analyses in most included studies. We calculated the HR from available data or Kaplan-Meier curves.
In conclusion, this meta-analysis suggested that CXCR4 overexpression was significantly associated to lymph node metastasis, distant metastasis, tumor stage and overall survival in lung cancer. CXCR4 might be a new prognostic biomarker, and it might become a new diagnostic and therapeutic target in lung cancer. Further studies are required to explore the molecular biology mechanism of CXCR4 and factors that result in significant heterogeneity in our meta-analysis.
Acknowledgements
This study was supported by National Natural Science Foundation of China (No. 81001113). The authors are most grateful to all the participants in this study.
Disclosure of conflict of interest
None.
References
- 1.Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer J Clin. 2013;1:11–30. doi: 10.3322/caac.21166. [DOI] [PubMed] [Google Scholar]
- 2.Nesbitt JC, Putnam JJ, Walsh GL, Roth JA, Mountain CF. Survival in early-stage non-small cell lung cancer. Ann Thorac Surg. 1995;2:466–472. doi: 10.1016/0003-4975(95)00169-l. [DOI] [PubMed] [Google Scholar]
- 3.Mareel M, Leroy A. Clinical, cellular, and molecular aspects of cancer invasion. Physiol Rev. 2003;2:337–376. doi: 10.1152/physrev.00024.2002. [DOI] [PubMed] [Google Scholar]
- 4.Busillo JM, Benovic JL. Regulation of cxcr4 signaling. Biochim Biophys Acta. 2007;4:952–963. doi: 10.1016/j.bbamem.2006.11.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kucia M, Jankowski K, Reca R, Wysoczynski M, Bandura L, Allendorf DJ, Zhang J, Ratajczak J, Ratajczak MZ. Cxcr4-sdf-1 signalling, locomotion, chemotaxis and adhesion. J Mol Histol. 2004;3:233–245. doi: 10.1023/b:hijo.0000032355.66152.b8. [DOI] [PubMed] [Google Scholar]
- 6.Burger JA, Kipps TJ. Cxcr4: a key receptor in the crosstalk between tumor cells and their microenvironment. Blood. 2006;5:1761–1767. doi: 10.1182/blood-2005-08-3182. [DOI] [PubMed] [Google Scholar]
- 7.Yang P, Liang SX, Huang WH, Zhang HW, Li XL, Xie LH, Du CW, Zhang GJ. Aberrant expression of cxcr4 significantly contributes to metastasis and predicts poor clinical outcome in breast cancer. Curr Mol Med. 2014;1:174–184. doi: 10.2174/1566524013666131121115656. [DOI] [PubMed] [Google Scholar]
- 8.Kim J, Mori T, Chen SL, Amersi FF, Martinez SR, Kuo C, Turner RR, Ye X, Bilchik AJ, Morton DL, Hoon DS. Chemokine receptor cxcr4 expression in patients with melanoma and colorectal cancer liver metastases and the association with disease outcome. Ann Surg. 2006;1:113–120. doi: 10.1097/01.sla.0000217690.65909.9c. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lee HJ, Jo DY. The role of the cxcr4/cxcl12 axis and its clinical implications in gastric cancer. Histol Histopathol. 2012;9:1155–1161. doi: 10.14670/HH-27.1155. [DOI] [PubMed] [Google Scholar]
- 10.Taichman RS, Cooper C, Keller ET, Pienta KJ, Taichman NS, McCauley LK. Use of the stromal cell-derived factor-1/cxcr4 pathway in prostate cancer metastasis to bone. Cancer Res. 2002;6:1832–1837. [PubMed] [Google Scholar]
- 11.Spano JP, Andre F, Morat L, Sabatier L, Besse B, Combadiere C, Deterre P, Martin A, Azorin J, Valeyre D, Khayat D, Le Chevalier T, Soria JC. Chemokine receptor cxcr4 and early-stage non-small cell lung cancer: pattern of expression and correlation with outcome. Ann Oncol. 2004;4:613–617. doi: 10.1093/annonc/mdh136. [DOI] [PubMed] [Google Scholar]
- 12.Zhang MX, Yu SY, Xia S. Effect of chemokine receptor CXCR4 on the invasion and metastasis capability in lung cancer. Acta Med Univ Sci Technol Huazhong. 2006;35:801–803. [Google Scholar]
- 13.Cai FY, Xin JB, Shi XL, Tian SJ. The expression of stromal-cell derived factor 1 (SDF-1), CXC chemokin receptor 4 (CXCR4) and their clinical significances in human non-small cell lung cancer. Cancer Res Prev Trea. 2006;33:578–580. [Google Scholar]
- 14.Suzuki M, Mohamed S, Nakajima T, Kubo R, Tian L, Fujiwara T, Suzuki H, Nagato K, Chiyo M, Motohashi S, Yasufuku K, Iyoda A, Yoshida S, Sekine Y, Shibuya K, Hiroshima K, Nakatani Y, Yoshino I, Fujisawa T. Aberrant methylation of cxcl12 in non-small cell lung cancer is associated with an unfavorable prognosis. Int J Oncol. 2008;1:113–119. [PubMed] [Google Scholar]
- 15.Wagner PL, Hyjek E, Vazquez MF, Meherally D, Liu YF, Chadwick PA, Rengifo T, Sica GL, Port JL, Lee PC, Paul S, Altorki NK, Saqi A. Cxcl12 and cxcr4 in adenocarcinoma of the lung: association with metastasis and survival. J Thorac Cardiovasc Surg. 2009;3:615–621. doi: 10.1016/j.jtcvs.2008.07.039. [DOI] [PubMed] [Google Scholar]
- 16.Iwakiri S, Mino N, Takahashi T, Sonobe M, Nagai S, Okubo K, Wada H, Date H, Miyahara R. Higher expression of chemokine receptor cxcr7 is linked to early and metastatic recurrence in pathological stage i nonsmall cell lung cancer. Cancer. 2009;11:2580–2593. doi: 10.1002/cncr.24281. [DOI] [PubMed] [Google Scholar]
- 17.Reckamp KL, Figlin RA, Burdick MD, Dubinett SM, Elashoff RM, Strieter RM. Cxcr4 expression on circulating pan-cytokeratin positive cells is associated with survival in patients with advanced non-small cell lung cancer. BMC Cancer. 2009;9:213. doi: 10.1186/1471-2407-9-213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Xiao TH, Ren Y, Chen XW, Liang LW. Expression and clinical significance of chemokine receptor CXCR4 in lung cancer and its corresponding lymph node metastasis. Clin J Med Off. 2009;37:408–410. [Google Scholar]
- 19.Minamiya YY, Saito H, Takahashi N, Ito M, Imai K, Ono T, Motoyama S, Ogawa J. Expression of the chemokine receptor cxcr4 correlates with a favorable prognosis in patients with adenocarcinoma of the lung. Lung Cancer. 2010;3:466–471. doi: 10.1016/j.lungcan.2009.07.015. [DOI] [PubMed] [Google Scholar]
- 20.Yao JF, Wang MZ, Wang P. Expression and clinical significancer of CXCR4 and HPA in non-small cell lung cancer cells. Hebei Med J. 2010;32:2162–2164. [Google Scholar]
- 21.Chen G, Wang Z, Liu XY, Liu FY. High-level cxcr4 expression correlates with brain-specific metastasis of non-small cell lung cancer. World J Surg. 2011;1:56–61. doi: 10.1007/s00268-010-0784-x. [DOI] [PubMed] [Google Scholar]
- 22.Otsuka S, Klimowicz AC, Kopciuk K, Petrillo SK, Konno M, Hao D, Muzik H, Stolte E, Boland W, Morris D, Magliocco AM, Bebb DG. Cxcr4 overexpression is associated with poor outcome in females diagnosed with stage IV non-small cell lung cancer. J Thorac Oncol. 2011;7:1169–1178. doi: 10.1097/JTO.0b013e3182199a99. [DOI] [PubMed] [Google Scholar]
- 23.Wang M, Chen GY, Song HT, Hong X, Yang ZY, Sui GJ. Significance of cxcr4, phosphorylated stat3 and vegf-A expression in resected non-small cell lung cancer. Exp Ther Med. 2011;3:517–522. doi: 10.3892/etm.2011.235. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Xi Y, Wang YR, Song Y. The study on expression of matrix CXC chemokin receptor 4 and microvessel density in non-small cell lung cancer. Clin J Lab Diagn. 2011;15:72–75. [Google Scholar]
- 25.Li RJ, Zhao LJ, Zhan ZL, Lv X, Gong LL, Wang P. Significance of expression of chemokine receptor and matrix metalloproteinase in small cell lung cancer. Zhonghua Yi Xue Za Zhi. 2012;92:532–535. [PubMed] [Google Scholar]
- 26.Zhou Z, Chen ZW, Yang XH, Shen L, Ai XH, Lu S, Luo QQ. Establishment of a biomarker model for predicting bone metastasis in resected stage iii non-small cell lung cancer. J Exp Clin Cancer Res. 2012;31:34. doi: 10.1186/1756-9966-31-34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Geng H, Xu ML. Expression of chemokine CXCL12/CXCR4 in non-small cell lung cancer. Tianjin Med J. 2012;40:203–205. [Google Scholar]
- 28.Hu TJ. Expression and its significance of CXCR4 and MMP-13 in non-small cell lung carcinoma. China Modern Doctor. 2012;50:47–49. [Google Scholar]
- 29.Wang SB, Zhou H, Li ZT, Cao WK, Han QZ. Expressions and significance of CXCR4 and CCR7 chemokine receptor in lung cancer. Chin J Gen Prac. 2012;10:1713–1715. [Google Scholar]
- 30.Zobair AA, Obeidy BF, Yang L, Yang C, Hui Y, Yu H, Zheng F, Yang G, Xie C, Zhou F, Zhou Y. Concomitant overexpression of egfr and cxcr4 is associated with worse prognosis in a new molecular subtype of non-small cell lung cancer. Oncol Rep. 2013;4:1524–1532. doi: 10.3892/or.2013.2254. [DOI] [PubMed] [Google Scholar]
- 31.Wang MZ, Bai F, Liu JK, Chen LH, Huang WF. Expression and clinical significance of CXCR4 and VEGF in non-small cell lung cancer (NSCLC) cells. Modern Onco. 2013;21:1989–1992. [Google Scholar]
- 32.Li Y, Shen Y, Miao Y, Luan Y, Sun B, Qiu X. Co-expression of upar and cxcr4 promotes tumor growth and metastasis in small cell lung cancer. Int J Clin Exp Pathol. 2014;7:3771–3780. [PMC free article] [PubMed] [Google Scholar]
- 33.Liu L, Lv Y, Wang Y, Ren XL. Expression and clinical significance of EGFR, HER2 and CXCR4 in non-small cell lung cancer. Prog Modern Biomed. 2014;14:1069–1073. [Google Scholar]
- 34.Ji T, Zhu SB, Dong YQ, Zhang XM, Liang JS, Yin GL. Clinical significance of TFPI-2 and CXCR4 expression in non-small cell lung cancer. Chin J Clin Healthc. 2014;17:288–290. [Google Scholar]
- 35.Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials. 2007;8:16. doi: 10.1186/1745-6215-8-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Liu CF, Liu SY, Min XY, Ji YY, Wang N, Liu D, Ma N, Li ZF, Li K. The prognostic value of cxcr4 in ovarian cancer: a meta-analysis. PLoS One. 2014;3:e92629. doi: 10.1371/journal.pone.0092629. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Zhang Z, Ni C, Chen W, Wu P, Wang Z, Yin J, Huang J, Qiu F. Expression of cxcr4 and breast cancer prognosis: a systematic review and meta-analysis. BMC Cancer. 2014;14:49. doi: 10.1186/1471-2407-14-49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Han M, Lv S, Zhang Y, Yi R, Huang B, Fu H, Bian R, Li X. The prognosis and clinicopathology of cxcr4 in gastric cancer patients: a meta-analysis. Tumour Biol. 2014;5:4589–4597. doi: 10.1007/s13277-013-1603-4. [DOI] [PubMed] [Google Scholar]
- 39.Balkwill FR. The chemokine system and cancer. J Pathol. 2012;2:148–157. doi: 10.1002/path.3029. [DOI] [PubMed] [Google Scholar]
- 40.Murphy PM. Chemokines and the molecular basis of cancer metastasis. N Engl J Med. 2001;11:833–835. doi: 10.1056/NEJM200109133451113. [DOI] [PubMed] [Google Scholar]
- 41.Kijima T, Maulik G, Ma PC, Tibaldi EV, Turner RE, Rollins B, Sattler M, Johnson BE, Salgia R. Regulation of cellular proliferation, cytoskeletal function, and signal transduction through cxcr4 and c-kit in small cell lung cancer cells. Cancer Res. 2002;21:6304–6311. [PubMed] [Google Scholar]
- 42.Orimo A, Gupta PB, Sgroi DC, Arenzana-Seisdedos F, Delaunay T, Naeem R, Carey VJ, Richardson AL, Weinberg RA. Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated sdf-1/cxcl12 secretion. Cell. 2005;3:335–348. doi: 10.1016/j.cell.2005.02.034. [DOI] [PubMed] [Google Scholar]
- 43.Li YM, Pan Y, Wei Y, Cheng X, Zhou BP, Tan M, Zhou X, Xia W, Hortobagyi GN, Yu D, Hung MC. Upregulation of cxcr4 is essential for her2-mediated tumor metastasis. Cancer Cell. 2004;5:459–469. doi: 10.1016/j.ccr.2004.09.027. [DOI] [PubMed] [Google Scholar]
- 44.Liu YL, Yu JM, Song XR, Wang XW, Xing LG, Gao BB. Regulation of the chemokine receptor cxcr4 and metastasis by hypoxia-inducible factor in non small cell lung cancer cell lines. Cancer Biol Ther. 2006;10:1320–1326. doi: 10.4161/cbt.5.10.3162. [DOI] [PubMed] [Google Scholar]
- 45.Bertolini G, Roz L, Perego P, Tortoreto M, Fontanella E, Gatti L, Pratesi G, Fabbri A, Andriani F, Tinelli S, Roz E, Caserini R, Lo VS, Camerini T, Mariani L, Delia D, Calabro E, Pastorino U, Sozzi G. Highly tumorigenic lung cancer cd133+ cells display stem-like features and are spared by cisplatin treatment. Proc Natl Acad Sci U S A. 2009;38:16281–16286. doi: 10.1073/pnas.0905653106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Burger M, Glodek A, Hartmann T, Schmitt-Graff A, Silberstein LE, Fujii N, Kipps TJ, Burger JA. Functional expression of cxcr4 (cd184) on small-cell lung cancer cells mediates migration, integrin activation, and adhesion to stromal cells. Oncogene. 2003;50:8093–8101. doi: 10.1038/sj.onc.1207097. [DOI] [PubMed] [Google Scholar]